International Scouting in Hockey

Evaluating European and International Prospects

Scouting international hockey prospects requires understanding different league contexts, playing styles, and development systems. European leagues emphasize different skills than North American hockey, making statistical adjustments crucial for fair evaluation.

Cross-League Prospect Comparison

import pandas as pd
import numpy as np
from sklearn.preprocessing import StandardScaler

# Load international prospect data
shl_prospects = pd.read_csv('shl_player_stats.csv')
liiga_prospects = pd.read_csv('liiga_player_stats.csv')
khl_prospects = pd.read_csv('khl_player_stats.csv')

# League quality adjustment factors
league_adjustments = {
    'SHL': 1.15,
    'Liiga': 1.12,
    'KHL': 1.10,
    'Allsvenskan': 0.95,
    'Mestis': 0.90,
    'MHL': 0.75,
    'Swiss-A': 1.05,
    'DEL': 1.00,
    'Czechia': 1.02
}

def calculate_adjusted_production(row):
    """Calculate league and age-adjusted production"""
    ppg = row['points'] / row['games_played']

    # League quality adjustment
    league_factor = league_adjustments.get(row['league'], 1.0)

    # Age adjustment (younger players get bonus)
    age_factor = 1 + (21 - row['age']) * 0.04

    # Ice time adjustment (limited ice time)
    if row['toi_per_game'] < 15:
        toi_factor = 1 + (15 - row['toi_per_game']) * 0.03
    else:
        toi_factor = 1.0

    adjusted_ppg = ppg * league_factor * age_factor * toi_factor

    return adjusted_ppg

# Combine all international prospects
all_prospects = pd.concat([shl_prospects, liiga_prospects, khl_prospects])

# Calculate adjusted metrics
all_prospects['adj_ppg'] = all_prospects.apply(
    calculate_adjusted_production, axis=1
)

# Calculate even-strength focus (important for NHL translation)
all_prospects['es_production_ratio'] = (
    all_prospects['es_points'] / all_prospects['points']
)

# Physical readiness score
def calculate_physical_readiness(row):
    """Assess physical NHL readiness"""
    score = 0

    # Size component (0-40 points)
    if row['height_cm'] >= 185 and row['weight_kg'] >= 90:
        score += 40
    elif row['height_cm'] >= 180 and row['weight_kg'] >= 85:
        score += 30
    elif row['height_cm'] >= 175 and row['weight_kg'] >= 80:
        score += 20
    else:
        score += 10

    # Age/development (0-30 points)
    if row['age'] >= 20:
        score += 30
    elif row['age'] >= 19:
        score += 20
    else:
        score += 10

    # League level (0-30 points)
    if row['league'] in ['SHL', 'Liiga', 'KHL']:
        score += 30
    elif row['league'] in ['Allsvenskan', 'Mestis', 'Swiss-A']:
        score += 20
    else:
        score += 10

    return score

all_prospects['physical_readiness'] = all_prospects.apply(
    calculate_physical_readiness, axis=1
)

# Create comprehensive prospect ranking
all_prospects['prospect_score'] = (
    all_prospects['adj_ppg'] * 30 +
    all_prospects['es_production_ratio'] * 20 +
    all_prospects['physical_readiness'] * 0.5
)

# Top international prospects
top_intl_prospects = all_prospects.sort_values(
    'prospect_score', ascending=False
).head(20)

print("=== Top International Hockey Prospects ===")
print(top_intl_prospects[[
    'name', 'age', 'league', 'points', 'games_played',
    'adj_ppg', 'es_production_ratio', 'physical_readiness', 'prospect_score'
]])

# Playing style analysis
def classify_playing_style(row):
    """Classify player type based on statistics"""
    goals_per_game = row['goals'] / row['games_played']
    assists_per_game = row['assists'] / row['games_played']

    if goals_per_game > assists_per_game * 1.3:
        return 'Shooter'
    elif assists_per_game > goals_per_game * 1.3:
        return 'Playmaker'
    else:
        return 'Balanced'

all_prospects['playing_style'] = all_prospects.apply(
    classify_playing_style, axis=1
)

# League comparison analysis
league_comparison = all_prospects.groupby('league').agg({
    'adj_ppg': 'mean',
    'es_production_ratio': 'mean',
    'physical_readiness': 'mean',
    'name': 'count'
}).rename(columns={'name': 'prospect_count'})

print("\n=== League Comparison ===")
print(league_comparison.sort_values('adj_ppg', ascending=False))

# Projection to NHL performance
def project_nhl_impact(row):
    """Project NHL impact timeline"""
    readiness_factors = []

    # Production factor
    if row['adj_ppg'] >= 0.8:
        readiness_factors.append('elite_production')
    elif row['adj_ppg'] >= 0.6:
        readiness_factors.append('strong_production')

    # Physical factor
    if row['physical_readiness'] >= 70:
        readiness_factors.append('nhl_ready_body')

    # Experience factor
    if row['games_played'] >= 50:
        readiness_factors.append('league_experienced')

    # Age factor
    if row['age'] >= 20:
        readiness_factors.append('mature')

    # Project timeline
    if len(readiness_factors) >= 4:
        return 'NHL Ready (0-1 year)'
    elif len(readiness_factors) >= 3:
        return 'Near Term (1-2 years)'
    elif len(readiness_factors) >= 2:
        return 'Development (2-3 years)'
    else:
        return 'Long Term (3+ years)'

top_intl_prospects['nhl_projection'] = top_intl_prospects.apply(
    project_nhl_impact, axis=1
)

print("\n=== NHL Readiness Projections ===")
print(top_intl_prospects[[
    'name', 'age', 'league', 'adj_ppg',
    'physical_readiness', 'nhl_projection'
]])

library(tidyverse)
library(scales)

# Load international prospect data
shl_prospects <- read_csv("shl_player_stats.csv")
liiga_prospects <- read_csv("liiga_player_stats.csv")
khl_prospects <- read_csv("khl_player_stats.csv")

# League adjustment factors
league_adjustments <- c(
  "SHL" = 1.15, "Liiga" = 1.12, "KHL" = 1.10,
  "Allsvenskan" = 0.95, "Mestis" = 0.90, "MHL" = 0.75,
  "Swiss-A" = 1.05, "DEL" = 1.00, "Czechia" = 1.02
)

# Calculate adjusted production
calculate_adjusted_production <- function(points, games, league, age, toi) {
  ppg <- points / games

  # League quality factor
  league_factor <- ifelse(league %in% names(league_adjustments),
                         league_adjustments[league], 1.0)

  # Age adjustment
  age_factor <- 1 + (21 - age) * 0.04

  # Ice time adjustment
  toi_factor <- ifelse(toi < 15, 1 + (15 - toi) * 0.03, 1.0)

  ppg * league_factor * age_factor * toi_factor
}

# Combine all prospects
all_prospects <- bind_rows(
  shl_prospects,
  liiga_prospects,
  khl_prospects
) %>%
  mutate(
    adj_ppg = calculate_adjusted_production(
      points, games_played, league, age, toi_per_game
    ),
    es_production_ratio = es_points / points
  )

# Physical readiness scoring
calculate_physical_readiness <- function(height, weight, age, league) {
  score <- 0

  # Size component
  score <- score + case_when(
    height >= 185 & weight >= 90 ~ 40,
    height >= 180 & weight >= 85 ~ 30,
    height >= 175 & weight >= 80 ~ 20,
    TRUE ~ 10
  )

  # Age component
  score <- score + case_when(
    age >= 20 ~ 30,
    age >= 19 ~ 20,
    TRUE ~ 10
  )

  # League component
  score <- score + case_when(
    league %in% c("SHL", "Liiga", "KHL") ~ 30,
    league %in% c("Allsvenskan", "Mestis", "Swiss-A") ~ 20,
    TRUE ~ 10
  )

  return(score)
}

all_prospects <- all_prospects %>%
  rowwise() %>%
  mutate(
    physical_readiness = calculate_physical_readiness(
      height_cm, weight_kg, age, league
    ),
    prospect_score = adj_ppg * 30 + es_production_ratio * 20 +
                    physical_readiness * 0.5
  ) %>%
  ungroup()

# Top prospects
top_intl_prospects <- all_prospects %>%
  arrange(desc(prospect_score)) %>%
  head(20)

cat("=== Top International Hockey Prospects ===\n")
print(top_intl_prospects %>%
  select(name, age, league, points, games_played, adj_ppg,
         es_production_ratio, physical_readiness, prospect_score))

# Playing style classification
all_prospects <- all_prospects %>%
  mutate(
    goals_per_game = goals / games_played,
    assists_per_game = assists / games_played,
    playing_style = case_when(
      goals_per_game > assists_per_game * 1.3 ~ "Shooter",
      assists_per_game > goals_per_game * 1.3 ~ "Playmaker",
      TRUE ~ "Balanced"
    )
  )

# League comparison
league_comparison <- all_prospects %>%
  group_by(league) %>%
  summarise(
    avg_adj_ppg = mean(adj_ppg, na.rm = TRUE),
    avg_es_ratio = mean(es_production_ratio, na.rm = TRUE),
    avg_physical = mean(physical_readiness, na.rm = TRUE),
    prospect_count = n()
  ) %>%
  arrange(desc(avg_adj_ppg))

cat("\n=== League Comparison ===\n")
print(league_comparison)

# Visualize prospect landscape
ggplot(all_prospects,
       aes(x = age, y = adj_ppg, color = league, size = physical_readiness)) +
  geom_point(alpha = 0.6) +
  scale_size_continuous(range = c(2, 8)) +
  labs(title = "International Hockey Prospect Landscape",
       subtitle = "Adjusted production vs age by league",
       x = "Age", y = "Adjusted Points Per Game",
       color = "League", size = "Physical Readiness") +
  theme_minimal() +
  theme(legend.position = "right")

# NHL projection analysis
project_nhl_timeline <- function(adj_ppg, physical_readiness,
                                games_played, age) {
  readiness_count <- 0

  if (adj_ppg >= 0.8) readiness_count <- readiness_count + 1
  if (adj_ppg >= 0.6) readiness_count <- readiness_count + 1
  if (physical_readiness >= 70) readiness_count <- readiness_count + 1
  if (games_played >= 50) readiness_count <- readiness_count + 1
  if (age >= 20) readiness_count <- readiness_count + 1

  case_when(
    readiness_count >= 4 ~ "NHL Ready (0-1 year)",
    readiness_count >= 3 ~ "Near Term (1-2 years)",
    readiness_count >= 2 ~ "Development (2-3 years)",
    TRUE ~ "Long Term (3+ years)"
  )
}

top_intl_prospects <- top_intl_prospects %>%
  rowwise() %>%
  mutate(
    nhl_projection = project_nhl_timeline(
      adj_ppg, physical_readiness, games_played, age
    )
  ) %>%
  ungroup()

cat("\n=== NHL Readiness Projections ===\n")
print(top_intl_prospects %>%
  select(name, age, league, adj_ppg, physical_readiness, nhl_projection))

Cultural and System Differences

European leagues develop players differently than North American systems. Swedish and Finnish leagues emphasize tactical awareness and two-way play, while Russian leagues on larger ice surfaces promote skill and creativity. Understanding these differences helps project NHL fit.